Forums

After digging out the credits for Headbuster, I've decided that searching and replacing over 50 kana characters is too much work to do manually.

I would like to know what you use to deal with Japanese text in games. Maxim, what did you do to generate normal 16-bit Japanese from the text in Phantasy Star, for example?

If it's in some standard 8-bit japanese codepage, you can open the data in Mozilla and change encoding settings. If it works you can then copy the text. It is a sursprisingly useful feature of web browsers.

Perhaps this might be of some use?

http://www.disgruntleddesigner.com/chrisc/HeadBusterEnglish/

I translated the game to English, including the credits, in 2005.

Quote

ASSISTANT PRODUCER
TOSHIRO TSUCHIDA
NORIHIKO YONESAKA

GAME DESIGNER
EIJI KOYAMA (WINDS)

CHIEF PROGRAMMER
KATSUTOSHI YAMAMOTO
TAKAHIRO MATSUZAWA

GRAPHIC DESIGNER
KAIROU

MUSIC COMPOSER
CUBE

PACKAGE DESIGNER
SATOSHI NAKAI (WINDS)

SPECIAL THANKS TO
YUKIHIRO HIGUCHI
MORIYAH

SALES PROMOTER
JUN MIMURA
YASUAKI YOKOE

EXECUTIVE PRODUCER
YASUMASA SHIRAKURA

idrougge wrote

I would like to know what you use to deal with Japanese text in games. Maxim, what did you do to generate normal 16-bit Japanese from the text in Phantasy Star, for example?

I didn't, as it happens, but I know how it was done. Basically, you need to define a mapping between (data in the ROM) and (Unicode, UTF-8 etc representation of the character). For PS, a "TBL" format file was used. (I think it's a fairly common format for ROM translation efforts - see http://www.zophar.net/utilities/texthack.html for some outdated listings.) Then it's just a matter of getting a script dumper to go through a list of ROM addresses and output the converted text looked up from the table.

So, to put it another way: there exist standard file formats for text dumping, avoiding the search/replace stage you mentioned.

In the long run, it's only worth it for cases where there's a lot of text to handle. For credits, it's probably quicker to type it in based on screenshots. Even if that does mean "typing" in the IME Pad.

ccovell wrote

Perhaps this might be of some use?

http://www.disgruntleddesigner.com/chrisc/HeadBusterEnglish/

Well, according to your readme, you had to do the work manually too, or did you?

Maxim wrote

In the long run, it's only worth it for cases where there's a lot of text to handle. For credits, it's probably quicker to type it in based on screenshots.

But if I had screenshots, I would have had this problem to begin with. But your hint put me in the right direction, I found a table translator for OSX on a ROM hacking site and now it's only a tenth of the original effort. Thanks!

Yup, I was quite disappointed with the sorry state of Table/Pointer generators. At least all the ones I could find:

myself wrote

I want to take the time now to point out deliberately that, although I researched and tried out
about 7 or 8 different "table/pointer/translation" software packages, all of them fell far short
of my expectations. Some were MS-DOS only, some crashed with some VB or Basic run-time error right
at the beginning, others mangled both input and output tables, and others seemed designed to work
with Super Famicom games only. So, I had to do it all by hand. That's quite disappointing.

Author	Message
idrougge Joined: 01 Jan 2005 Posts: 360 Location: Stockholm, Sweden	8-bit Japanese text encoding Posted: Mon Mar 24, 2008 11:25 pm
	After digging out the credits for Headbuster, I've decided that searching and replacing over 50 kana characters is too much work to do manually. I would like to know what you use to deal with Japanese text in games. Maxim, what did you do to generate normal 16-bit Japanese from the text in Phantasy Star, for example?

Bock Site Admin Joined: 08 Jul 2001 Posts: 8663 Location: Paris, France	Posted: Mon Mar 24, 2008 11:41 pm
	If it's in some standard 8-bit japanese codepage, you can open the data in Mozilla and change encoding settings. If it works you can then copy the text. It is a sursprisingly useful feature of web browsers.

ccovell Joined: 26 Dec 2004 Posts: 374 Location: Japan	Posted: Tue Mar 25, 2008 1:23 pm
ccovell Joined: 26 Dec 2004 Posts: 374 Location: Japan	Perhaps this might be of some use? http://www.disgruntleddesigner.com/chrisc/HeadBusterEnglish/ I translated the game to English, including the credits, in 2005. Quote ASSISTANT PRODUCER TOSHIRO TSUCHIDA NORIHIKO YONESAKA GAME DESIGNER EIJI KOYAMA (WINDS) CHIEF PROGRAMMER KATSUTOSHI YAMAMOTO TAKAHIRO MATSUZAWA GRAPHIC DESIGNER KAIROU MUSIC COMPOSER CUBE PACKAGE DESIGNER SATOSHI NAKAI (WINDS) SPECIAL THANKS TO YUKIHIRO HIGUCHI MORIYAH SALES PROMOTER JUN MIMURA YASUAKI YOKOE EXECUTIVE PRODUCER YASUMASA SHIRAKURA

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14759 Location: London	Posted: Tue Mar 25, 2008 2:13 pm
	idrougge wrote I would like to know what you use to deal with Japanese text in games. Maxim, what did you do to generate normal 16-bit Japanese from the text in Phantasy Star, for example? I didn't, as it happens, but I know how it was done. Basically, you need to define a mapping between (data in the ROM) and (Unicode, UTF-8 etc representation of the character). For PS, a "TBL" format file was used. (I think it's a fairly common format for ROM translation efforts - see http://www.zophar.net/utilities/texthack.html for some outdated listings.) Then it's just a matter of getting a script dumper to go through a list of ROM addresses and output the converted text looked up from the table. So, to put it another way: there exist standard file formats for text dumping, avoiding the search/replace stage you mentioned. In the long run, it's only worth it for cases where there's a lot of text to handle. For credits, it's probably quicker to type it in based on screenshots. Even if that does mean "typing" in the IME Pad.

idrougge Joined: 01 Jan 2005 Posts: 360 Location: Stockholm, Sweden	Posted: Tue Mar 25, 2008 8:12 pm
	ccovell wrote Perhaps this might be of some use? http://www.disgruntleddesigner.com/chrisc/HeadBusterEnglish/ Well, according to your readme, you had to do the work manually too, or did you? Maxim wrote In the long run, it's only worth it for cases where there's a lot of text to handle. For credits, it's probably quicker to type it in based on screenshots. But if I had screenshots, I would have had this problem to begin with. But your hint put me in the right direction, I found a table translator for OSX on a ROM hacking site and now it's only a tenth of the original effort. Thanks!

ccovell Joined: 26 Dec 2004 Posts: 374 Location: Japan	Posted: Wed Mar 26, 2008 1:54 am
ccovell Joined: 26 Dec 2004 Posts: 374 Location: Japan	Yup, I was quite disappointed with the sorry state of Table/Pointer generators. At least all the ones I could find: myself wrote I want to take the time now to point out deliberately that, although I researched and tried out about 7 or 8 different "table/pointer/translation" software packages, all of them fell far short of my expectations. Some were MS-DOS only, some crashed with some VB or Basic run-time error right at the beginning, others mangled both input and output tables, and others seemed designed to work with Super Famicom games only. So, I had to do it all by hand. That's quite disappointing.

Forums

View topic - 8-bit Japanese text encoding